Proposed is a multi-level study is to determine the association between features of the urban environment and sexual and drug use risk factors for HIV infection among habitual drug users. Recent evidence about the growing role of sex risk behavior in the transmission of HIV, even among habitual drug using populations, suggests that more comprehensive studies of the determinants of both sex and drug risks, and their relationship to HIV transmission, are important. Although there are at present several multi-site intervention studies that aim to minimize individual high risk sex and drug-use behavior, surprisingly little systematic research has explored the relation between features of the individual's environment, her/his risk behavior, and disease morbidity (sexually transmitted diseases and HIV) while controlling for relevant individual-level covariates. This study will examine the independent and interactive effect of features of the urban physical environment (e.g., population density, public transportation, housing quality) and of features of the social environment (e.g., violence, policing, income distribution, neighborhood disadvantage) as they relate to sexual partner dynamics, sexual and drug use risk factors for HIV infection, and to HIV and STD prevalence. To address the study aims, we will recruit 2,000 habitual drug users (including injectors and non-injectors) from eight New York City neighborhoods using multi-stage sampling over a four year period to allow for temporal trend analysis. Features of the urban environment will be assessed using publicly available archival information (e.g., New York City Housing and Vacancy Survey, Mayor's Management Report). Hierarchical multivariate modeling will be used to determine the relations between features of the urban environment and individual risk behavior, HIV, and STDs. We will also model the associations between changing features of the city and current behavior. Identification of the key features of the urban environment that shape risk behavior can guide structural interventions, increase the efficiency of individual-level behavioral interventions, and identify features of urban neighborhoods that could suggest population vulnerability to rapid transmission of HIV. [unreadable] [unreadable]